We were reading some lucid dreaming research papers by Stephen Laberge from Stanford (http://www.asdreams.org/journal/articles/laberge5-3.htm). These papers suggested that it might be possible to induce lucid dreams by delivering light cues during REM sleep, so we decided to use the Oura ring and some custom hardware to test this out.
What it does
It helps you lucid dream by shining a light pattern into your eyes when you're in REM sleep. This will cue you to let you know that you're in a dream, making you lucid.
How we built it
There were three main components to our project:
Machine learning: We used python's numpy library to run data analysis and predictions based on the sample Oura data. We were able to figure out what time the user usually has REM sleep and how long it lasts.
iOS app: We used an iOS app to connect to our machine learning web server and display the pattern of light at the right time.
The Dream Machine: We couldn't get any VR headsets or smart lights, so we had to design our own hardware from the ground up. We used abandoned cardboard boxes, tape, scissors and random materials we found around us to build this brand new innovative device that is optimized for shining a light pattern into the user's eyes.
Challenges we ran into
- Limited materials required us to design our own hardware. We had never done this before but decided to try it out.
- The Oura API was a bit difficult to use at first be eventually we figured it out.
- The ring currently doesn't give real time data so we had to use machine learning to predict the right time for REM sleep
- Lucid Dreaming is a sometimes controversial realm and it is hard to get concrete information, so we had to use bleeding-edge research from Stanford to figure out the design of our product.
Accomplishments that we're proud of
- Our 'Dream Machine' is frequently mistaken for google cardboard
- Used Oura API for the first time
- Used machine learning
- Made a beautiful iOS app
- Democratized lucid dreaming by making open source hardware
What we learned
- We learned about lucid dreams from the extensive research we did and papers we read
- How to use machine learning to make accurate predictions
- How to design hardware with the user in mind
What's next for Dream Machine
- Open source the design so anyone at home can make their own 'Dream Machine' out of cardboard.
- Integrate with real-time data when Oura updates their API.